Model atmospheric distillation

This chapter serves as a guide to model atmospheric distillation section of the crude distillation unit. We provide relevant process, operational and modeling details to model the atmospheric column. We also discuss methods to estimate missing data for model development We provide step-by-step instructions to model a particular column in Aspen HYSYS. We discuss how to validate the model predictions with plant data and how to use the model to perform industrially useful case studies. 

Several of the commercial simulation programs offer preconfigured complex column rigorous models for petroleum fractionation. These models include charge heaters, several side strippers, and one or two pump-around loops. These fractionation column models can be used to model refinery distillation operations such as crude oil distillation, vacuum distillation of atmospheric residue oil, fluidized catalytic cracking (FCC) process main columns, and hydrocracker or coker main columns. Aspen Plus also has a shortcut fractionation model, SCFrac, which can be used to configure fractionation columns in the same way that shortcut distillation models are used to initialize multicomponent rigorous distillation models. 

Each of the processes shown in Figure 2.1 is quite extensive and can be surprisingly complex in an integrated refinery. In this work, we limit the scope by presenting brief summaries of the unit and how to deal with each unit in a modeling context Our specific focus is how to model the atmospheric distillation section. 

The first step in the building the atmospheric distillation unit is entering the composition of the crude in order to generate the necessary hypothetical components for model. For the purposes of this simulation, we will consider the crude assays given in Table 2.5 to Table 2.8. It is important to remember that that we may have to remove extraneous details from the distillation curve to avoid unusual column behavior. We use the TB P distillation, density distribution and overall bulk density to define this system in Figure 2.14. 

Before we can use the model to study different operating scenarios and perform case studies, we must ensure that the model matches the base-line column conditions and operating profiles. For the atmospheric distillation column, the important operating profile measurements are ... 

After we validate the the model predictions with plant data, we would typically like to use the model to predict new operating scenarios or perform experiments that are too costly or otherwise prohibitive in actual atmospheric distillation. Refiners spend significant effort to develop models but they are rarely used again. Often times, the users neglect these models while the real column operation continues to change. Thus, when users actually run models, the predictions are far removed from process reality. The simple way to avoid this model stagnation is to use... 

The isotopic composition of the vapor and condensate are simply offset at any given height in the atmosphere by the appropriate equilibrium fractionation factor and hence the derivative of the isotopic composition for the vapor and condensate with respect to elevation are identical as described below by Equation (1). Open system distillation, as modeled by Rayleigh condensation, removes the condensate as it condenses from the vapor leaving the isotopic composition of the residual vapor progressively depleted in lsO and 2H. We use Q = -In(p/ps)... 

This natural process by which dissolved and/or particulate surface-active materials end up in the atmosphere has been modeled and studied in the laboratory. As summarized by Detwiler and Blanchard (ref. 46), tests in suspensions of bacteria (ref. 76,96,97), latex spheres (ref. 98), dyes (ref. 99), and in sea water and river water (ref. 96,100,101) have demonstrated successful transfer of all manner of surface-active material from the bulk fluid, or the bulk interface, to the droplets ejected when bubbles burst. (This situation can be pictured as an extension of the common industrial adsorptive-bubble-separation process (ref. 102) into a third dimension or phase — the atmosphere.) Further laboratory tests with various tap waters, distilled waters, and salt solutions have shown that no water sample was ever encountered that did not contain at least traces of surface-active material (ref. 46). 

Well over 50 large-scale EO model-based RTO applications have been deployed for petroleum refining processes. These model applications have been deployed in petroleum refineries Liporace et al., 2009 Camolesi et al., 2008 Mudt et al., 1995, both on separation units (crude atmospheric and vacuum distillation units) and on reactor units (including fluidized catalytic crackers (FCC), gasoline reformers, and hydrocrackers). 

Model the dissolution of quartz and K-feldspar (adularia) over time. Are the parameters temperature and C02 partial pressure of any importance Within the key word RATES use the BASIC program from the data set PHREEQC.dat. The calculation is done with distilled water (pH = 7, pE = 12) as a batch reaction over a time span of 10 years in 100 time steps at a temperatures of 5 °C and of 25 °C and at C02 partial pressures of 0.035 Vol% (atmosphere) and of 0.7 Vol% (soil). Calculate also the kinetics of the dissolution with 0.035 Vol% C02 and 25 °C for a period of 10 minutes. 

Figures 10 and 11 show some examples of model predictions of the feedstock characterization of crude distillation residues. Figure 10 compares model predictions with the experimental distillation curves of three Arabian atmospheric residues. Figure 11 shows the model ability in predicting the aromatic carbon content and the H/C of different feeds in comparison with some NMR data. A more detailed description and discussion of this residue characterization is reported elsewhere (Bozzano et al., 1995, 1998).

In preparing all solutions, the distilled, deionized water was boiled to remove C02 prior to its use. All samples were then kept sealed to prevent any absorption of C02 during equilibration. For the Sephadex, a 3-4-day period was sufficient to achieve equilibrium with the IRC-50, however, the equilibrium interval exceeded 2 weeks. At equilibrium, pH and pNa measurements at 25° 0.1 °C were made with the system under a nitrogen atmosphere. All pH measurements were made with a glass electrode using a Radiometer Model 4 pH meter. A sodium-ion selective electrode was used to measure pNa. A saturated calomel electrode was used as the reference electrode. 

---